1. Field of the Invention
The present invention is directed to a nucleic acid binding compound comprising N8- or C8-linked purine bases or structurally related heterocycles, a compound useful for the preparation of such compound, a binding product of this nucleic acid binding compound with a nucleic acid, a method for the determination of a nucleic acid using said compound, and several uses of 8-linked purine bases and structurally related heterocycles.
2. Description of the Related Art
The synthesis of some C8-glycosylated purine derivates is described in Czech. Chem. Comm. (Vol 34, p. 247), Heterocycl. Chem. (Vol 12, p. 111) and Journal of Medicinal Chemistry (Vol 33, p. 2750-2755). Examples of N8-glycosylated purine derivates are shown in Helv. Chim. Acta (Vol. 72, p. 1527), Zh. Org. Khim. (Vol 12, p. 1131) and Helv. Chim. Acta (Vol 76, p. 2184-2193). However, there is no disclosure to use these non-natural bases as universal bases capable of base pairing with each of the four natural bases when incorporated in nucleic acids.
Some examples of 8-aza-7-deazaadenine coupled to a sugar moiety at its N8-position are known in the art. In Helv. Chim. Acta (Vol 72, p. 868-881), there is disclosed the synthesis of 8-aza-7-deazaadenine N8-(β-D-2′-deoxyribofuranoside). The synthesis of N6-methylated-8-aza-7-deazaadenine N8-(β-D-2′-deoxyribofuranoside) is shown in Helv. Chim. Acta (Vol 71, p. 1813-1823). However, there is no disclosure showing the advantages using such nucleotides as universal bases.
In Biochemistry (Vol 15, p. 1005-1015), there is disclosed the synthesis of 8-aza-7-deazaadenine N8-(β-D-ribofuranosyl) (A*). Also shown is the enzymatic polymerization of ADP and A*DP in the presence of polynucleotide phosphorylase. However, there is no disclosure on the synthesis of oligonucleotides having a defined sequence containing other bases than those two. The hybridization behavior of these polymers is not described. Also no hint is given that these modified bases can be used in hybridization methods.
The present invention is particularly useful for nucleic acid hybridization methods, for example in nucleic acid determinations in analytics, especially in the field of health care. It can also be used for in-vitro-mutagenesis. It can also be applied in the pharmaceutical field, for example in oligonucleotides to be used as antisense oligonucleotides.
Nucleic acids have been found to be useful analytes for the determination of the presence or absence of genes or microorganisms in human body fluid, food or environment. Especially since the introduction of nucleic amplification techniques like the Polymerase Chain Reaction (PCR, see U.S. Pat. No. 4,683,202) the determination of nucleic acids is widely used, because of their good sensitivity.
The amplified nucleic acids can be determined using a variety of different techniques, dependent from the particular purpose. Most assays require the use of a probe which is either immobilized or immobilizable or is labeled by attachment of one or more reporter groups. A reporter group has the characteristics to be itself capable to be determined or it can be reacted with reagents that make the probe determinable via said reporter group. Thus, for example, probes that are labeled by reporter groups can be determined, as can be hybrids that contain the probe and a nucleic acid to be determined.
Regarding the optional amplification step as well as the probe hybridization step the hybridization behavior of the oligonucleotides used as primers and probes is very important. Dependent from the particular purpose, there are many proposals to include modified or non-natural heterocyclic groups instead of natural nucleobases in such oligonucleotides in order to improve the hybridization method.
An example of such an non-natural group is 7-deaza-dGTP which, when introduced into a nucleic acid replacing dGTP reduces band compression in sequencing gels (EP-B-0 286 028).
Nucleic acid determination is largely based on the fact, that the natural canonical bases A, C, G and T are only able to base pair with only one of the four natural bases. However the different stability of A paired with T compared to G paired with C could be a problem, because therefore the melting point of an hybrid not only depends on the length of the probe oligonucleotide but also on its GC-content.
In methods for screening of a nucleic acid in the case that only the amino acid sequence of the accompanying protein is available there is another problem. In that case the nucleic acid sequence can be deduced from the amino acid sequence. However the redundancy of the genetic code leads to a multitude of different sequences. Mixtures of oligonucleotides that take this redundancy into account must be synthesized and used for screening the potential DNA or RNA candidates.
An alternative approach uses less discriminatory base analogues. Such base is defined as a universal or ambiguous base, the corresponding nucleosides are named universal or ambiguous nucleosides. A universal base is one which pairs with some or all of the natural bases without major discrimination. Hypoxanthine for which the term universal base was coined is widely used as universal base, because it can form hydrogen bonds with all four natural bases (Y. Takahashi, K. Kato, Y. Hayashizaki, T. Wakabayashi, E. Ohtsuka, S. Matsuki, M. Ikehara, K. Matsubara, Proc. Natl. Acad. Sci. USA 1985, 1931-1935). When 2′-deoxyinosine (see FIG. 1, 1), the nucleoside of hypoxanthine, is incorporated into a 12-mer duplex opposite each of the four natural nucleosides they show a wide range of thermal stability represented by Tm-value differences of 15° C. (F. Seela, K. Mittelbach, Nucleosides, Nucleotides 1999, 18, 425-441). This shows that hypoxanthine does not base pair equally with the other bases. Furthermore, hypoxanthine being used in PCR primers behaves like guanine. Other universal nucleosides are the compounds P (see FIG. 1, 2) and K (see FIG. 1, 3) (P. Kong Thoo Lin, D. M. Brown, Nucleic Acids Res. 17, 10373-10383; D. M. Brown, P. Kong Thoo Lin, Carbohydrate Res., 216, 129-139). They were chosen as their amino-imino tautomeric forms of their bases are much more to unity than the normal bases. The nucleoside P can form Watson-Crick base pairs with the purine nucleosides dA and dG while the nucleoside K forms base pairs with dT and dC.
Apart from the nucleosides forming hydrogen bonds there are a number of universal residues which possess no hydrogen bonding capabilities. The electronic charge distribution of such heterocycles are good candidates for base stacking abasic sites (see FIG. 1, 4) (T. A. Milican, G. A. Mock, M. A. Chaunncy, T. P. Patel, M. A. W. Eaton, J. Gunning, S. D. Cutbush, S. Neidle, J. Mann, Nucleic Acids Res. 1984, 12, 7435-7453), and replacement of the natural bases by 5-nitroindol (see FIG. 1, 5) (D. Lokes, D. M. Brown, Nucleic Acids Res. 1994, 22, 4039-4043), 3-nitropyrrol (see FIG. 1, 6) (D. E. Bergstrom, P. Zhang, W. Travis Johnson, Nucleic Acids Res. 1997, 25, 1935-1942) has been used for such work.
All proposals known now have some disadvantages. Therefore, there is still a need to provide nucleic acid binding compounds like primers and probes containing universal bases capable to better base pair with each of the four natural bases.